Earlier work by the P.I. (Newman 1965), showed that the anti-malaria campaign carried out in Sri Lanka during the late 1940s had large effects on rates of mortality and of fertility. However, the presently estimated behavior of fertility rates with respect to malaria prevalance contains many puzzles. For example, pre-program fertility rates were uncorrelated, and post-program fertility rates strongly correlated, with the pre-program prevalance of malaria. The present proposal seeks a better explanation of these puzzles. Specifically, it is proposed to examine one or more of the recently developed models of human fertility (e.g., Bongaarts 1976) and explore systematically the influence that malarial prevalance had on the various parameters that enter that model. Special attention will be paid to the role of fetal and infant mortality, using partly the approach of Bourgeois-Pichat (B-P) to the latter phenomenon. As a part of this research, new methods of estimation of the B-P parameters will be used. An attempt will be made to estimate from the Sri Lanka data all the parameters of the appropriate form of the fertility model used, for both pre- and post-program periods. An essential preliminary to this investigation will be further scrutiny of the data on fertility and infant mortality, by district, for Sri Lanka from 1930 through 1960.